ACEEE Int. J. on Transportation and Urban Development, Vol. 01, No.

01, Apr 2011

The Effect of Climatic Conditions of Various Regions of Iran on Pavement Fatigue Cracking
S.M.Nasimifar 1,M.R.Pouranian2, and M.Azadi 3 M. Sc. Civil Engineering, Sharif University of Technology, Tehran, Iran Email: nasimifar@yahoo.com 2 M. Sc. Civil Engineering, Sharif University of Technology, Tehran, Iran Email: mrp325@gmail.com 3 Professor, Islamic Azad University, Qazvin Branch, Qazvin, Iran Email: azadi@aut.ac.ir
1

AbstractIn this research, climatic data for various regions of Iran were obtained from meteorological organizations in their original formats and then were converted to ICM 1 input files for MEPDG2 software by algorithms. The major aim of this research is to assess the environmental effects on fatigue cracking, so more than 300 analyses were conducted on a specific pavement with climatic data from various regions of Iran, since pavement temperature and humidity profile have important effects on alligator and longitudinal cracking, and specific pavements dont have the same performance in all regions. Lastly, useful solutions will be recommended for decreasing the deleterious effects of climatic conditions on fatigue cracking Index TermsMEPDG model, Climatic parameters, Fatigue cracking I.
BACKGROUND

The pavement design procedures presented in the American Association of State Highway Officials (AASHTO) Guide for Design of Pavement Structures [1] require the use of mechanical properties for the asphalt concrete, base course, and soil subgrade. The stiffness of the soil subgrade and base materials are represented by the resilient modulus, R M which replaces the empirical soil support value used in the earlier design guides. A sensitivity analysis of the AASHTOs design equation showed that the resilient modulus of the unbound materials has the most pronounced effect on the structural number (SN) of flexible pavements [2]. Since the behavior of unbound base materials is similar to that of coarsegrained subgrade materials, the effects of moisture content changes are similar. Typically, the resilient modulus of unbound materials is determined in the laboratory in accordance with AASHTO T294 [3] under conditions of maximum dry density and optimum water content. Although pavement subgrades are usually compacted close to optimum water content and maximum dry density during construction, seasonal variations in water content or degree of saturation occur. Most fine-grained soils exhibit a decrease in modulus as the water content is increased, leading to increased

deflections in the pavement subgrade. Coarse-grained materials may experience this change, depending upon the amount of fine grained particles present. In general, an increased deflection in the subgrade leads to a decrease in pavement design life ([4];[5]; [6]). The structural capacity of flexible pavements is heavily influenced by the stiffness of the asphalt concrete layer. The asphalt concrete stiffness is a function of temperature and varies throughout the day as well as through the year. Asphalt concrete temperature can affect the structural performance of the pavement in two ways. The stiffness of asphalt concrete is directly related to pavement temperature; the stiffness decreases as the temperature increases. A decrease in asphalt concrete stiffness results in higher stresses being transmitted to the base and subgrade. Most base and subgrade materials are stress-dependent. Typically, granular materials are stiffer at higher stress levels and cohesive soils are weaker at higher stress levels. Therefore, the asphalt concrete temperature indirectly affects the behavior of the base and subgrade. This variation in structural capacity with time and temperature means that pavement damage does not occur uniformly throughout the year. In order for pavement engineers to design flexible pavements efficiently, temperature effects must be considered in the design process. In order to account for temperature effects, relationships between temperature and asphalt concrete stiffness must be developed and the asphalt temperature must be determined. The asphalt temperature can be determined from direct measurement or from correlations with weather data. II. INTRODUCTION Recent researches have shown that environmental conditions have a significant effect on the performance of flexible pavements. Interaction between environmental parameters, pavement materials, and the loading condition is very complicated. Factors such as precipitation, temperature, freeze-thaw cycles and ground water table depth play a key role in defining the moisture and temperature profile in the pavement structure. These parameters can significantly affect the pavement layer and subgrade properties and hence, its load capacity [7]. 27

2011 ACEEE

DOI: 01.IJTUD.01.01.27

ACEEE Int. J. on Transportation and Urban Development, Vol. 01, No. 01, Apr 2011 Thanks to the rise of rapid calculation tools, climatic effects a parameter that has been disregarded in the past because of long calculations and a lack of suitable tools for analysis has been marked. Pavements expose various climatic conditions in the design life. Climatic efficacy has been evaluated in AASHTO 2004 comprehensively. The key conceptual differences between AASHTO 2004 and AASHTO 1993 are summarised in table 1[8]. Environmental effects can be subdivided into two factors: external factors, such as precipitation, temperature, freezethaw cycles and ground water table depth; and internal factors, such as the susceptibility of the pavement materials to moisture and freeze-thaw damage, the drain ability of the paving layers, the infiltration potential of the pavement, et cetera, define the extent to which the pavement will react to the applied external environmental conditions [7]. Iran has a variety of climatic conditions. The evaluation of Iran climatic data from 1968 to 2008 indicates that there is a difference about 20(C) in mean annual temperature between Ardebil and Bandarabbas and there is a difference about 130 cm in mean annual rainfall between Rasht and Yazd. These differences are indicative of various climates. For the calculation of the environmental condition effects on the performance of the asphalt pavements, AASHTO 2004 design software that is nominated MEPDG has been utilised. In this software, climatic parameters are the basic input parameters and it is impossible to analyse the pavement without suitable climatic data. In this research, suitable climatic data files that are useful in MEPDG format were extracted from various climatic stations of Iran. Then, one type pavement that is more prevalent in Iran was selected and the effects of the environmental conditions of various regions of Iran on the fatigue cracking were evaluated. III. PREPARATION OF INPUT DATA FOR ANALYSIS The first step of analysis is the appointment of suitable assumptions. Since the purpose of this research is to evaluate the relationship between the environment and fatigue cracking, so the same assumptions have been proposed for all climatic conditions and for various inputs such as materials pavement properties, layer thickness and traffic loading. Then we compared the obtained results to each other. Finally, for sensitivity analyses, we changed the type of binder and asphalt layer thickness. The assumptions of these analyses are presented by the following paragraph IV.1. TRAFFIC For traffic input data in all analyses, interstate roads assumptions that exist in software were used. 4.2. Pavement structure A cross-section of a simple conventional flexible pavement was used in this study. The structure is a fourlayer pavement system including a single asphalt concrete layer, two unbound granular bases, subbase layers and a subgrade. Figure 1 shows the pavement structure that was used in the study. The unbound granular properties were the same for all computer runs. Table 2 summarizes the input properties of the unbound layers. Two different asphalt concretes were used in this study. Table 3 shows asphalt mixture properties. The difference between two type mixes is the viscosity gradient of binder. Mix type I was used in all analysis and mix type II was used for sensitivity analyses. Figure 2 illustrates the dynamic modulus of two mixes types under the same climatic conditions. This figure simply implies the comparison between the two mixes in which mix type II has a less dynamic modulus in contrast with the mix type I. V. PAVEMENT ANALYSES FOR VARIOUS CLIMATIC CONDITIONS Hence, the pavement model has been produced (per last sections preamble) and climatic files of 30 centres of Iran state have been applied for analyses.

Figure 1- Section of the asphalt pavement in the modeling

2011 ACEEE

DOI: 01.IJTUD.01.01.27

28

ACEEE Int. J. on Transportation and Urban Development, Vol. 01, No. 01, Apr 2011

Temperature and humidity profiles play key roles on the performance of pavement, particularly with fatigue cracking [9]. Air temperature, cloud cover and wind speed have been characterized with pavement temperature [7 and 10]. The Comparison between maximum temperatures in the middle of asphalt layer in basic climatic stations is shown in Table 4. According to this table, maximum and minimum temperatures of the asphalt layer were experienced in Ahvaz and Ardebil respectively. Table 5 shows the comparison between mean annual rainfalls in various climatic stations. In this table, Rasht is the first rainy state among the other states. The results in this section were obtained from analyses on the pavement that are presented in section 2 and spot mix type and with the same traffic loading for all analyses. Ground water table depth was considered 3 meters in all analyses. The analyses are only different in the input climatic file proportionate at design site. In figure 3, the centres of states are sorted by the alligator cracking, and increase from left to right. In this figure, stations that experience higher temperatures have high alligator cracking. The result of this modelling is evident in figure 4. In this diagram the effect of mean annual air temperature on the alligator cracking has been assessed.

Figure 2- Comparison of dynamic modulus of the asphalt mixtures in the same climatic data

Based on this figure, it does not separate the effect on the other environmental parameters, such as humidity and other parameters like wind speed and cloud cover, that play a role in temperature profile. Thus, this diagram is not absolutely ascendant; nonetheless it can be found that the amount of alligator cracking will increase with the increase of the mean annual air temperature at the design site. Increasing the humidity causes the subgrade modulus to decrease and hence tensile stress increases at the bottom of the asphalt layer. Thus, alligator cracking rises. This fact can be seen in the situation of Rasht in figure 3 and table 4. Although this city is a member of the first 10 cold stations according to table 4, because of high rainfall, it experiences high alligator cracking

2011 ACEEE

29

DOI: 01.IJTUD.01.01.27

ACEEE Int. J. on Transportation and Urban Development, Vol. 01, No. 01, Apr 2011

Figure 4- Correlation between alligator cracking and mean annual air temperature

VI. SENSITIVITY ANALYSIS TO ENVIRONMENTAL PARAMETERS AND

STRUCTURAL PROPERTIES

In figure 5, the centres of states are sorted by the higher longitudinal cracking from left to right. In this figure, the stations that experiment higher temperature have high longitudinal cracking. Unlike the alligator cracking, in longitudinal cracking, decreasing subgrade modulus causes the tensile stress at the top of the asphalt layer to decrease. So increasing the humidity resulted in a decreasing subgrade modulus and a decreasing longitudinal cracking. The Rasht situation in figure 3 and figure 5 shows this fact.

According to the analyses, increasing the temperature and humidity has negative effects on the performance of pavement. The objective of this major component of the overall design procedure sensitivity study is to investigate how the prediction of fatigue cracking is affected by magnitude variations of several different key input climatic parameters. To evaluate the effects of temperature variations, we changed the daily air temperature by 5 C & 10C in 3 stations (Ahvaz, Tehran and Ardebil) and analyses were performed for periods of 40 months in each station. Input parameters that have been described in section 3 are the same and all input parameters except air temperature have been fixed for any analyses. In this way, the effect of the air temperature variations was specified on the performance of pavement. The most important effect of the air temperature variations is the effect on the asphalt layer. Increasing the air temperature causes the maximum temperature of the asphalt layer to increase and thereby reduces the asphalt layers modulus.

Figure 5- longitudinal cracking versus the asphalt layer temperature and humidity

Figure 6 shows a correlation between top to down fatigue damage that is in relation to longitudinal cracking and maximum average monthly temperature surface.
Figure 3- Alligator cracking versus the asphalt layer temperature and humidity

2011 ACEEE

30

DOI: 01.IJTUD.01.01.27

ACEEE Int. J. on Transportation and Urban Development, Vol. 01, No. 01, Apr 2011 Fatigue damage depends on the asphalt layer modulus and critical tensile strain [11]. Decreasing the asphalt layer modulus (due to increasing air temperature) causes fatigue damage to increase. In order to evaluate the effect of humidity pavement profile variations on the fatigue cracking, analyses were done by fixing all input parameters variations but the ground water table depth was changed by 1.5, 3, 4.5 and 6 meters. Figure 7 shows longitudinal cracking variations in Tehran station versus the variations of ground water depth. Despite the decreasing dynamic modulus of the asphalt layer, which results in a rise in longitudinal cracking, increasing the subgrade modulus causes the generation of surface tensile strains in the asphalt layer and increases the longitudinal cracking. As the ground water table depth increases, the subgrade modulus increases due to decreasing humidity, and longitudinal cracking increases. Despite longitudinal cracking, when subgrade modulus decreases in alligator cracking, more tensile stress occurs in the bottom of the asphalt layer and alligator cracking increases. Figure 8 shows alligator cracking variations versus the variations of ground water table depth. To evaluate the effect of AC thickness on the fatigue cracking, analyses were done for 160 months period and 5 different AC thicknesses with mix type in Tehran station. Other input parameters were the same according to section 2. Figure 9 illustrates variations of the longitudinal cracking and alligator cracking with AC layer thickness. The most important conclusion of figure 9 is that for good performance, the proper thickness of AC layers must be either as thin as practical or thick as possible. The fundamental reasoning behind the results is the utilization of a mechanistic approach to pavement design [12].

Figure 7- Longitudinal cracking variations versus the variations of the ground water table depth

Since in obtained results one model has been used in all analyses, the effect of the pavement structural specifications in modelling such as bitumen applied in asphalt mixture can also be used for sensitivity analyses. For this purpose, the second type of asphalt mixture that was described in section 3 was used for analysis. As expected, changing the process performance of pavement with climatic parameters is independent of model type, but the intensity level of these changes will be different, a fact which will be further discussed. In all sensitivity analyses, other model parameters such as traffic and unbound layers of the same equivalent of default have been considered in section2. To evaluate the binder effect, mixture type II for Tehran station was used for sensitivity temperature analyses. These analyses were investigated in two thicknesses of 10 and 20 cm for asphalt layer. Also the temperature was changed by 5 and 10 C and the result will be assessed in 5 temperature points. Figure 10 shows variations of longitudinal cracking versus the mean annual air temperature.

Figure 6- Correlation between top-down fatigue damage and the maximum average monthly temperature surface

Figure 8- Alligator cracking variations versus the variations of the ground water table depth

2011 ACEEE

31

DOI: 01.IJTUD.01.01.27

ACEEE Int. J. on Transportation and Urban Development, Vol. 01, No. 01, Apr 2011 performance but while AC thickness is 20cm, mix type I (high dynamic modulus) has slightly better performance. From two figures 10 & 11 we can generally conclude that, in the thin AC layers (thickness of less than 10 cm), using a asphalt mixture with low dynamic modulus causes lower fatigue cracking and in thick AC layers (thickness of more than 15 cm), this reduction happens in a high dynamic modulus of mixture. VII. CONCLUSION This research describes the effects of the environmental parameters (such as temperature and humidity) in different regions of Iran on the fatigue cracking for prevalent pavement in Iran. Pavements are exposed to different climatic conditions in their design life. Thus, there will not be the same performance in different climatic regions. Therefore, the appropriate analyses are carried out and their results are summarised as follows: 1. When mean annual air temperature rises and as a result of it the asphalt layer temperature increases, the dynamic modulus of asphalt mixture decreases in pavements with any structural specifications and fatigue damages increase. 2. Decreasing the subgrade modulus due to increasing moisture causes the tensile stress at the bottom of the asphalt layer to rise and then alligator cracking (bottom up damage) increases. But longitudinal cracking develops from the top down (top/down damage). Therefore, decreasing the subgrade modulus due to increasing moisture causes both the tensile stress at the top of the asphalt layer and the longitudinal cracking to decrease. Thus, moisture has different effects on the alligator and longitudinal cracking and the higher ground water table depth and therefore higher subgrade modulus causes the lower alligator cracking and the higher longitudinal cracking to occur 3. Critical AC layer thickness is 10 cm for alligator cracking and 15 cm for longitudinal cracking. 4. To reduce the deleterious effects of pavement temperature increments on the alligator cracking, the use of an asphalt mixture with a low dynamic modulus is recommended. Also, using the suitable drainage system for reduction of alligator cracking due to increasing the moisture is recommended. 5. To reduce negative effects of increasing pavement temperature on the longitudinal cracking, for thin asphalt layer (less than 10 cm), using an asphalt mixture with low dynamic modulus is recommended, and for a thick asphalt layer (more than 15 cm), using an asphalt mixture with higher dynamic modulus is recommended.

Figure9- Thickness effect of the AC layer on the fatigue cracking

According to section 2, the features of two types of mixes are the same, only in the second one there is bitumen Pen 85100 instead of bitumen Pen 60-70. This variation cause the dynamic modulus and average tensile strength of two mixes to change. Therefore, Mixed type II has lower dynamic modulus and a higher tensile strength in comparison with mixed type I. Tensile strength has a role of thermal cracking variations and dynamic modulus effects the fatigue cracking and rutting. The features of two mixed types are presented in table 3 Figure 10 indicates that longitudinal cracking in 10 cm asphalt layer is less than the thickness of 20cm. Another main point of these sensitivity analyses is that in 10cm AC layer (thin AC layer), the best AC mixture is one that exhibits low stiffness with Master Curve (type mix II). When the mixture becomes stiffer, the amount of longitudinal cracking due to top-down fatigue fracture increases as well. Thus, for 20cm AC layer (thick AC layer) in which the AC mix stiffness increases, the amount of surface longitudinal cracking decreases. Figure 11 shows variations of alligator cracking versus the mean annual air temperature. According to this figure, the performance of a thin AC layer is different from a thick AC layer and mix type has different effects. When AC thickness is 10 cm, mix type II (low dynamic modulus) shows a better

Figure10- MAAT effect on the longitudinal cracking

2011 ACEEE

32

DOI: 01.IJTUD.01.01.27

ACEEE Int. J. on Transportation and Urban Development, Vol. 01, No. 01, Apr 2011
[5] R. P. Elliot and S. I. Thornton, Resilient Modulus and AASHTO Pavement Design. Transportation Research Record 1196 (1988). [6] C. L. Monismith, Analytically-Based Asphalt Pavement Design and Rehabilitation-Theory to Practice (1962-1992), TRB Distinguished Lecture. Transportation Research Record 1354 5-25 (1992). [7] AASHTO. (2004). Guide for the Design of Pavement Structures, Part 2 Chapter 3, Environmental Effects (draft), American Association of State Highway Officials, Washington, D.C. [8] Carvalho, R.L, (2006). Mechanistic-Empirical Design of Flexible Pavements: A Sensitivity Study. University of Maryland, in partial fulfillment of the requirements for the degree of Master of Science, Advisory committee: Dr. Charles W. Schwartz, Professor Deborah J. Goodings, Dr. Dimitrios G. Goulias. [9] AASHTO. (2004). Guide for the Design of Pavement Structures, Appendix II-1, Calibration of Fatigue Cracking Models for Flexible Pavements (draft), American Association of State Highway Officials, Washington, D.C. [10] Birgisson, B, J. Ovik and D.E. Newcomb. (2000) Analytical Predictions of Seasonal Variation in Flexible Pavements at the Mn/ Road Site. the 79th Annual Meeting of the Transportation Research Board. Washington, D. C. [11] AASHTO. (2004). Guide for the Design of Pavement Structures, Part 3 Chapter 3, Design of New and Reconstructed Flexible Pavements (draft), American Association of State Highway Officials, Washington, D.C.

Figure11- MAAT effect on the alligator cracking

REFERENCES
[1] AASHTO. AASHTO Guide for the Design of Pavement Structures. 403. 1993.Washington, D.C., American Association of State Highway Officials. [2] R. L. Baus and J. A. Fogg, AASHTO Flexible Pavement Design Equation Study. Journal of Transportation Engineering 115, 559564 (1989). A 54 [3] AASHTO. AASHTO Interim Method of Test for Resilient Modulus of Unbound Granular Base/Subbase Materials and Subgrade Soils - SHRP Protocol P46. 1182. 1992. Washington, D.C., American Association of State Highway Officials. [4] M. R. Thompson and R. P. Elliot, ILLI-PAVE - Based Response Algorithms for Design of Conventional Flexible Pavements. Transportation Research Record 1043 50-57 (1985).

[12] Walubita, L. F. and MFC Van de ven (2000). Stresses and Strains in Asphalt-Surfacing Pavements. South African Transport Conference, South Africa.

2011 ACEEE

33

DOI: 01.IJTUD.01.01.27